Introduction

Most eukaryotic genes carry protein-coding exons that are separated by non-coding introns1,2. Pre-mRNA splicing is performed by the spliceosome, a large ribonucleoprotein complex comprised of five small nuclear ribonucleoproteins (snRNPs U1, U2, U4, U5 and U6) and a large number of associated proteins3,4. The size of introns ranges from 13 to over 300,000 nucleotides5,6. Sufficient evidence suggest that intronic sequences not only determine the splicing pattern7, but also have regulatory functions in gene expression8. Although most known regulatory sequences including the conserved GT and AG located at the beginning and the end of introns, respectively, an A at the branch point and a pyrimidine tract in spliceosome-binding and intron splicing are located in introns9, exonic sequences play an important role in accurate splicing as well10,11,12. The average size of exons is approximately 130 nucleotides in vertebrates and 180 nucleotides in plants13. Studies have showed that exons with less than 51 nucleotides may cause exon skipping and exons that are too small in size may hinder the recognition of adjacent spliceosome binding14,15,16,17. However, internal micro-exons with less than 25 nucleotides have been identified in different eukaryotic organisms by sequencing and computational analyses18,19. The smallest naturally available exon that has been experimentally characterized so far has 3 nucleotides16,20. Here we report the identification of a single-nucleotide exon in Arabidopsis.

Results

APC11 cDNA in GenBank is mis-annotated

APC11 (At3g05870) is a single-copy gene in the genome of Arabidopsis thaliana21. Current annotation predicts that APC11 has three exons and two introns and its coding sequence (CDS) contains 261 nucleotides, producing a polypeptide with 87 amino acids (AAs)21. However, sequencing of APC11 cDNA performed in this study has identified only one CDS with 252 nucleotides (highlighted in red; Fig. 1), encoding a polypeptide with 84 AAs. The discrepancy was partially caused by the inclusion of 10 nucleotides from the first intron to the exon in previous annotation (highlighted in blue; Fig. 1).

Figure 1
figure 1

The genomic sequence of APC11.

The coding region of APC11, in which putative exons are highlighted in red and capital, introns are denoted in black and lower case and the putative branch point “a” is highlighted in purple. A333 is the putative single-nucleotide exon. Conserved intron-exon splicing sequences “gt” and “ag” are underlined and in lower case. Start and stop codons are underlined and in capital. The mis-annotated exonic sequence in GenBank is highlighted in blue.

Full size image

Further, alignment of the cDNA obtained with the APC11 genomic sequence revealed a single-nucleotide A inserted into the cDNA. The mysterious A is not in continuity with the CDS in the genomic region. The insertion is absolutely required for in-frame APC11 translation. Re-sequencing of the APC11 genomic DNA extracted from both Col-0 and Ler ecotypes confirmed that the genomic sequence available in the GenBank of National Center for Biotechnology Information (NCBI) is correct, while its cDNA annotated is wrong. We therefore speculate that the extra A may originate from a single-nucleotide exon located in the intron between the previously annotated first and second exons. Within the assigned 422-nucleotide intronic sequence we identified a putative A (designated as A333 in Fig. 1), surrounded by GT and AG, located 333 nucleotides after the upstream exon-intron junction. A putative branch point A was detected 44 nucleotides upstream of the A333 (highlighted in purple; Fig. 1).

A333 is a functional single-nucleotide exon

To test whether A333 indeed represents a single-nucleotide exon, six constructs with nucleus-localized APC11-SV40-GFP fusion proteins expressed under the control of the cauliflower mosaic virus (CaMV) 35S promoter were made: 1) gAPC11-nGFP: the 839-nucleotide APC11 genomic sequence, with its stop codon deleted, in-frame fused with a SV40-GFP reporter gene; 2) cAPC11-nGFP: a 252-nucleotide APC11 cDNA, with its stop codon deleted, in-frame fused with the same SV40-GFP; 3) gAPC11(A > T)-nGFP: the same as gAPC11-nGFP except the A333 was substituted by a T, which is expected to produce a cDNA with T333 if the A333 is indeed a single-nucleotide exon; 4) gAPC11(A > G)-nGFP: the A333 in gAPC11-nGFP was substituted by a G to determine whether nucleotide types affect the splicing; 5) gAPC11(A > TT)-nGFP: the A333 in gAPC11-nGFP was substituted by TT, which shall cause a TT substitution in the APC11 cDNA and a frame shift in APC11 translation, leading to disappearance of GFP fluorescence; and 6) gAPC11(-A)-nGFP: A333 in gAPC11-nGFP was deleted, which shall produce a cDNA without A333, leading to a frame-shift in APC11 translation and disappearance in GFP fluorescence (Fig. 2a). These constructs were introduced into A. thaliana mesophyll protoplasts individually using a polyethylene glycol (PEG)-mediated transfection22 for in vivo transcriptional and translational assays.

Figure 2
figure 2

In vivo transcriptional and translational assays in Arabidopsis and rice protoplasts.

(a) Constructs generated for transient assays. The CaMV 35S promoter was used to drive the expression of APC11 cDNA, genomic or different substitution constructs fused with a nucleus-localized SV40-GFP reporter gene. Boxes in orange, cyan and grey indicate three previously identified exons in APC11. The black lines indicate introns and the A333 is shown as red vertical bars. (b) Alignment of APC11 cDNA produced in transgenic Arabidopsis or rice protoplasts. Identical nucleotides are shaded. (c) Examinations of GFP fluorescence in Arabidopsis protoplasts transfected with constructs illustrated in a. Note that GFP signals are detected only in protoplasts transfected with nGFP, cAPC11-nGFP, gAPC11-nGFP, gAPC11(A > T)-nGFP or gAPC11(A > G)-nGFP. Scale bar = 10 μm for all photos in c.

Full size image

cDNAs were prepared from RNAs extracted from protoplasts transfected with different fusion constructs to examine their splicing patterns. Afterwards, APC11-nGFP cDNAs were amplified from individual cDNAs by polymerase chain reaction (PCR) using a forward APC11 primer and a reverse GFP primer (Supplementary Table S1) and sequenced. Results obtained showed that, when either cAPC11-nGFP or gAPC11-nGFP was used, a sequence identical to APC11 cDNA was produced. Interestingly, substitutions of A333 by T [gAPC11(A > T)-nGFP], G [gAPC11(A > G)-nGFP] or TT [gAPC11(A > TT)-nGFP] led to T, G or TT substitutions in the cDNA, respectively (Fig. 2b). Further, deletion of A333 made in gAPC11(-A)-nGFP led to production of a cDNA without the A.

Detections of GFP fluorescence were used to define the translation of different fusion constructs. When examined under a confocal microscope after twelve-hour incubations, nucleus-localized GFP fluorescence was observed in protoplasts transfected with either cAPC11-nGFP, gAPC11-nGFP, gAPC11(A > T)-nGFP or gAPC11(A > G)-nGFP, suggesting that in-frame GFP translations were achieved in protoplasts transfected with these constructs. In contrast, no GFP fluorescence was detected when either gAPC11(A > TT)-nGFP or gAPC11(-A)-nGFP was used (Fig. 2c), indicating that the substitution of the A333 by TT or deletion of the A333 impaired the translation of these fusion constructs. These results confirmed that A333 in the APC11 is a functional single-nucleotide exon.

Splicing of the single-nucleotide exon is mostly conserved in rice

We then addressed whether the processing capability of the single-nucleotide exon is conserved in rice (Oryza sativa, var. Zhonghua 11), a remotely related monocotyledonous species. APC11 in rice has two paralogs, OsAPC11-1 (Os03g0302700) and OsAPC11-2 (Os07g0411101), both of them lack an intron. Protoplasts prepared from 14-day-old etiolated rice seedlings were used to perform in vitro transcriptional assay using above-mentioned six constructs (Fig. 2a). Sequencing of APC11-nGFP cDNAs amplified from rice protoplasts showed that, when either cAPC11-nGFP or gAPC11-nGFP was used in transfections, the intact APC11 cDNA produced from the same splicing patterns as those in Arabidopsis protoplasts were detected (Fig. 2b). Similarly, T or G substitutions were detected in cDNA isolated from protoplasts transfected with gAPC11(A > T)-nGFP or gAPC11(A > G)-nGFP, respectively (Fig. 2b). A cDNA without A333 and consequently a frame-shift, was detected in protoplasts transfected with gAPC11(-A)-nGFP. These results suggest that protoplasts of rice can splice the single-nucleotide exon accurately and effectively as those from Arabidopsis. However, it is interesting to note that, when gAPC11(A > TT)-nGFP was used, the splicing was incorrect. Additional 56 nucleotides from the first intron were incorporated into the cDNA, leading to a frame-shift in the translation of gAPC11(A > TT)-nGFP, suggesting that the substitution of A333 by TT has caused an altered splicing pattern in rice, which was not observed in Arabidopsis.

Discussion

Pre-mRNA splicing is essential in gene expression in eukaryotic organisms since most of their genes contain multiple copies of non-coding introns interspersed between exons. Precise removal of introns ensures the accurate production of proteins. Exons in pre-mRNA can be spliced either constitutively or alternatively: the former generates a single splicing product across all cell types and developmental stages in which the gene is expressed and the latter produces a variety of mRNAs by splicing from the same gene in different arrangements to generate protein diversity9,23. How these intronic sequences are removed effectively and accurately is still largely unknown, given the fact that the sizes of introns and exons varies tremendously5,6,13.

The average size of internal exons in most eukaryotic organisms is from 130 to 180 nucleotides13. Although it has been proposed that exons with less than 51 nucleotides may hinder the recognition of adjacent spliceosome binding, causing exon skipping14,15,16,17, micro-exons with less than 25 nucleotides have been identified in different eukaryotic organisms by sequencing and computational analyses18,19. For example, extensive studies have been performed in a 9-nucleotide constitutive micro-exon in the potato intertase gene and a 6-nucleotide constitutive micro-exon from the chicken cTNT gene24,25. The potato invertase gene carries an exon with 9 nucleotides. When 8 of these 9 nucleotides were deleted, the artificial 1-nucleotide exon was skipped in 33% transcripts produced. When this 9-nucleotide exon was replaced by a 6-nucleotide exon from the chicken cTNT gene, over 50% of the transcripts produced skipped or mis-spliced the exon24. Another recent study in animal and human brains have identified a whole set of genes carrying evolutionally conserved micro-exons, often with the numbers of multiples of three nucleotides, which are involved in modulating interaction domains of neural proteins through alternative splicing20. It is plausible that different regulatory mechanisms are implicated in splicing introns flanking a normal exon or a micro-exon.

Three models have been proposed to explain how pre-mRNA splicing is achieved. The “intron definition” model states that, for introns with moderate sizes, the splicing reaction occurs by pairing of the splice sites at two ends of an intron to remove the introns3,7. The “exon definition” model is proposed to explain the phenomenon that, for short exons separated by a large intervening intron, attaching a 5′ splice site downstream of the second exon in a two-exon splicing substrate greatly enhances the splicing of the upstream intron in vitro3,16. A “recursive splicing” model, which is proposed recently to explain the removal of large introns successively in several steps using intronic ratchet points26,27,28,29. In this study, we identified a constitutive single-nucleotide exon in Arabidopsis. In vitro transcriptional and translational assays performed in protoplasts showed that splicing of introns around this exon can be achieved accurately in both Arabidopsis and rice. We also demonstrated that nucleotide types, either purine or pyrimidine, have no effect on splicing of introns around the single-nucleotide exon. Given the fact that spliceosomes are very large in size30, it is very unlikely that the exon definition model could be used to explain the splicing of two introns flanking such a single-nucleotide exon. The intron definition model is more plausible, although it is very unlikely that two introns flanking the single-nucleotide exon could be spliced simultaneously. A combined intron definition and recursive splicing model might be applicable to explain the splicing of introns flanking the single-nucleotide exon, to allow two flanking introns to be removed one after another. Consistent with this hypothesis, it has been reported that in the potato invertase gene the splicing of introns surrounding the 9-nucleotide exon occurs recursively in two steps: the second intron was removed before the first one24. Further studies are needed to discriminate these possibilities and to identify regulatory sequences involved in intron splicing around the single-nucleotide exon.

In summary, although how widely such single-nucleotide exons are present in eukaryotic genomes remains to be investigated, the discovery of the functional single-nucleotide exon undoubtedly has significant impact on genome annotation in the future.

Materials and Methods

Plant materials

Arabidopsis thaliana plants (ecotypes Col-0 and Ler) were grown at 21 °C in a growth room with 16 h of light (100 μmol photons m−2sec−1) per day.

Constructs

SV40-GFP was amplified from pPLV0431. The full-length APC11 cDNA (cAPC11) was amplified from cDNA prepared from Col-0 seedlings using reverse transcription polymerase chain reaction (RT-PCR) and APC11 genomic DNA (gAPC11) was amplified from Col-0 or Ler genomic DNA. SV40-GFP and either cAPC11 or gAPC11 were ligated simultaneously into p326-cGFP digested with XbaI (NEB, USA) and KpnI (NEB, USA) using a one-step cloning assay32 to produce p35S:cAPC11-nGFP or p35S:gAPC11-nGFP, respectively. To generate p35S:nGFP, p35S:gAPC11-nGFP was digested by XbaI and ligated with T4 DNA ligase (NEB, USA). For A333 substitutions, point mutations were introduced to p35S:gAPC11-nGFP using the primers listed in Supplementary Table S1 to produce APC11(A > T), APC11(A > G), APC11(A > TT) or APC11(-A).

Protoplast transfection

For protoplast transient expressions, well-expanded leaves from 4-week-old Arabidopsis plants (Col-0) were chosen and the assays were performed as previously described22. For protoplast transient expression in rice, seeds (Oryza sativa, var. Zhonghua 11) were germinated on half-strength MS basal salts medium and cultured in the dark at 26 °C for 10 to 12 days before protoplasts were isolated and assays were performed as in Arabidopsis except the Macerozyme R-10 was replaced by Macerozyme RS (Yakult, Japan).

Microscopic analyses

To examine the expression of GFP in transfected protoplasts, a confocal laser scanning microscope (FV1000MPE, Olympus, Japan) equipped with 488 nm excitation laser was used.

RNA extraction and RT-PCR

Total RNA was isolated from transfected Arabidopsis or rice protoplasts using the Plant Total RNA Purification Kit (GeneMark, China), reverse-transcribed using the FastQuant RT Kit (TIANGEN, China) and sequenced.

Additional Information

How to cite this article: Guo, L. and Liu, C.M. A single-nucleotide exon found in Arabidopsis. Sci. Rep. 5, 18087; doi: 10.1038/srep18087 (2015).